1 /*- 2 * SPDX-License-Identifier: BSD-3-Clause 3 * 4 * Copyright (c) 1989, 1993 5 * The Regents of the University of California. All rights reserved. 6 * 7 * This code is derived from software contributed to Berkeley by 8 * Case Larsen. 9 * 10 * Redistribution and use in source and binary forms, with or without 11 * modification, are permitted provided that the following conditions 12 * are met: 13 * 1. Redistributions of source code must retain the above copyright 14 * notice, this list of conditions and the following disclaimer. 15 * 2. Redistributions in binary form must reproduce the above copyright 16 * notice, this list of conditions and the following disclaimer in the 17 * documentation and/or other materials provided with the distribution. 18 * 3. Neither the name of the University nor the names of its contributors 19 * may be used to endorse or promote products derived from this software 20 * without specific prior written permission. 21 * 22 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 23 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 24 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 25 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 26 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 27 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 28 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 29 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 30 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 31 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 32 * SUCH DAMAGE. 33 */ 34 35 #ifndef lint 36 static const char copyright[] = 37 "@(#) Copyright (c) 1989, 1993\n\ 38 The Regents of the University of California. All rights reserved.\n"; 39 #endif /* not lint */ 40 41 #ifndef lint 42 #endif /* not lint */ 43 44 #include <sys/capsicum.h> 45 46 #include <capsicum_helpers.h> 47 #include <ctype.h> 48 #include <err.h> 49 #include <errno.h> 50 #include <getopt.h> 51 #include <limits.h> 52 #include <locale.h> 53 #include <nl_types.h> 54 #include <stdint.h> 55 #include <stdio.h> 56 #include <stdlib.h> 57 #include <string.h> 58 #include <termios.h> 59 #include <unistd.h> 60 #include <wchar.h> 61 #include <wctype.h> 62 63 static int Dflag, cflag, dflag, uflag, iflag; 64 static int numchars, numfields, repeats; 65 66 /* Dflag values */ 67 #define DF_NONE 0 68 #define DF_NOSEP 1 69 #define DF_PRESEP 2 70 #define DF_POSTSEP 3 71 72 static const struct option long_opts[] = 73 { 74 {"all-repeated",optional_argument, NULL, 'D'}, 75 {"count", no_argument, NULL, 'c'}, 76 {"repeated", no_argument, NULL, 'd'}, 77 {"skip-fields", required_argument, NULL, 'f'}, 78 {"ignore-case", no_argument, NULL, 'i'}, 79 {"skip-chars", required_argument, NULL, 's'}, 80 {"unique", no_argument, NULL, 'u'}, 81 {NULL, no_argument, NULL, 0} 82 }; 83 84 static FILE *file(const char *, const char *); 85 static wchar_t *convert(const char *); 86 static int inlcmp(const char *, const char *); 87 static void show(FILE *, const char *); 88 static wchar_t *skip(wchar_t *); 89 static void obsolete(char *[]); 90 static void usage(void); 91 92 int 93 main (int argc, char *argv[]) 94 { 95 wchar_t *tprev, *tthis; 96 FILE *ifp, *ofp; 97 int ch, comp; 98 size_t prevbuflen, thisbuflen, b1; 99 char *prevline, *thisline, *p; 100 const char *ifn, *errstr;; 101 cap_rights_t rights; 102 103 (void) setlocale(LC_ALL, ""); 104 105 obsolete(argv); 106 while ((ch = getopt_long(argc, argv, "+D::cdif:s:u", long_opts, 107 NULL)) != -1) 108 switch (ch) { 109 case 'D': 110 if (optarg == NULL || strcasecmp(optarg, "none") == 0) 111 Dflag = DF_NOSEP; 112 else if (strcasecmp(optarg, "prepend") == 0) 113 Dflag = DF_PRESEP; 114 else if (strcasecmp(optarg, "separate") == 0) 115 Dflag = DF_POSTSEP; 116 else 117 usage(); 118 break; 119 case 'c': 120 cflag = 1; 121 break; 122 case 'd': 123 dflag = 1; 124 break; 125 case 'i': 126 iflag = 1; 127 break; 128 case 'f': 129 numfields = strtonum(optarg, 0, INT_MAX, &errstr); 130 if (errstr) 131 errx(1, "field skip value is %s: %s", errstr, optarg); 132 break; 133 case 's': 134 numchars = strtonum(optarg, 0, INT_MAX, &errstr); 135 if (errstr != NULL) 136 errx(1, "character skip value is %s: %s", errstr, optarg); 137 break; 138 case 'u': 139 uflag = 1; 140 break; 141 case '?': 142 default: 143 usage(); 144 } 145 146 argc -= optind; 147 argv += optind; 148 149 if (argc > 2) 150 usage(); 151 152 ifp = stdin; 153 ifn = "stdin"; 154 ofp = stdout; 155 if (argc > 0 && strcmp(argv[0], "-") != 0) 156 ifp = file(ifn = argv[0], "r"); 157 cap_rights_init(&rights, CAP_FSTAT, CAP_READ); 158 if (caph_rights_limit(fileno(ifp), &rights) < 0) 159 err(1, "unable to limit rights for %s", ifn); 160 cap_rights_init(&rights, CAP_FSTAT, CAP_WRITE); 161 if (argc > 1) 162 ofp = file(argv[1], "w"); 163 else 164 cap_rights_set(&rights, CAP_IOCTL); 165 if (caph_rights_limit(fileno(ofp), &rights) < 0) { 166 err(1, "unable to limit rights for %s", 167 argc > 1 ? argv[1] : "stdout"); 168 } 169 if (cap_rights_is_set(&rights, CAP_IOCTL)) { 170 unsigned long cmd; 171 172 cmd = TIOCGETA; /* required by isatty(3) in printf(3) */ 173 174 if (caph_ioctls_limit(fileno(ofp), &cmd, 1) < 0) { 175 err(1, "unable to limit ioctls for %s", 176 argc > 1 ? argv[1] : "stdout"); 177 } 178 } 179 180 caph_cache_catpages(); 181 if (caph_enter() < 0) 182 err(1, "unable to enter capability mode"); 183 184 prevbuflen = thisbuflen = 0; 185 prevline = thisline = NULL; 186 187 if (getline(&prevline, &prevbuflen, ifp) < 0) { 188 if (ferror(ifp)) 189 err(1, "%s", ifn); 190 exit(0); 191 } 192 tprev = convert(prevline); 193 194 tthis = NULL; 195 while (getline(&thisline, &thisbuflen, ifp) >= 0) { 196 if (tthis != NULL) 197 free(tthis); 198 tthis = convert(thisline); 199 200 if (tthis == NULL && tprev == NULL) 201 comp = inlcmp(thisline, prevline); 202 else if (tthis == NULL || tprev == NULL) 203 comp = 1; 204 else 205 comp = wcscoll(tthis, tprev); 206 207 if (comp) { 208 /* If different, print; set previous to new value. */ 209 if (Dflag == DF_POSTSEP && repeats > 0) 210 fputc('\n', ofp); 211 if (!Dflag) 212 show(ofp, prevline); 213 p = prevline; 214 b1 = prevbuflen; 215 prevline = thisline; 216 prevbuflen = thisbuflen; 217 if (tprev != NULL) 218 free(tprev); 219 tprev = tthis; 220 thisline = p; 221 thisbuflen = b1; 222 tthis = NULL; 223 repeats = 0; 224 } else { 225 if (Dflag) { 226 if (repeats == 0) { 227 if (Dflag == DF_PRESEP) 228 fputc('\n', ofp); 229 show(ofp, prevline); 230 } 231 show(ofp, thisline); 232 } 233 ++repeats; 234 } 235 } 236 if (ferror(ifp)) 237 err(1, "%s", ifn); 238 if (!Dflag) 239 show(ofp, prevline); 240 exit(0); 241 } 242 243 static wchar_t * 244 convert(const char *str) 245 { 246 size_t n; 247 wchar_t *buf, *ret, *p; 248 249 if ((n = mbstowcs(NULL, str, 0)) == (size_t)-1) 250 return (NULL); 251 if (SIZE_MAX / sizeof(*buf) < n + 1) 252 errx(1, "conversion buffer length overflow"); 253 if ((buf = malloc((n + 1) * sizeof(*buf))) == NULL) 254 err(1, "malloc"); 255 if (mbstowcs(buf, str, n + 1) != n) 256 errx(1, "internal mbstowcs() error"); 257 /* The last line may not end with \n. */ 258 if (n > 0 && buf[n - 1] == L'\n') 259 buf[n - 1] = L'\0'; 260 261 /* If requested get the chosen fields + character offsets. */ 262 if (numfields || numchars) { 263 if ((ret = wcsdup(skip(buf))) == NULL) 264 err(1, "wcsdup"); 265 free(buf); 266 } else 267 ret = buf; 268 269 if (iflag) { 270 for (p = ret; *p != L'\0'; p++) 271 *p = towlower(*p); 272 } 273 274 return (ret); 275 } 276 277 static int 278 inlcmp(const char *s1, const char *s2) 279 { 280 int c1, c2; 281 282 while (*s1 == *s2++) 283 if (*s1++ == '\0') 284 return (0); 285 c1 = (unsigned char)*s1; 286 c2 = (unsigned char)*(s2 - 1); 287 /* The last line may not end with \n. */ 288 if (c1 == '\n') 289 c1 = '\0'; 290 if (c2 == '\n') 291 c2 = '\0'; 292 return (c1 - c2); 293 } 294 295 /* 296 * show -- 297 * Output a line depending on the flags and number of repetitions 298 * of the line. 299 */ 300 static void 301 show(FILE *ofp, const char *str) 302 { 303 304 if ((!Dflag && dflag && repeats == 0) || (uflag && repeats > 0)) 305 return; 306 if (cflag) 307 (void)fprintf(ofp, "%4d %s", repeats + 1, str); 308 else 309 (void)fprintf(ofp, "%s", str); 310 } 311 312 static wchar_t * 313 skip(wchar_t *str) 314 { 315 int nchars, nfields; 316 317 for (nfields = 0; *str != L'\0' && nfields++ != numfields; ) { 318 while (iswblank(*str)) 319 str++; 320 while (*str != L'\0' && !iswblank(*str)) 321 str++; 322 } 323 for (nchars = numchars; nchars-- && *str != L'\0'; ++str) 324 ; 325 return(str); 326 } 327 328 static FILE * 329 file(const char *name, const char *mode) 330 { 331 FILE *fp; 332 333 if ((fp = fopen(name, mode)) == NULL) 334 err(1, "%s", name); 335 return(fp); 336 } 337 338 static void 339 obsolete(char *argv[]) 340 { 341 int len; 342 char *ap, *p, *start; 343 344 while ((ap = *++argv)) { 345 /* Return if "--" or not an option of any form. */ 346 if (ap[0] != '-') { 347 if (ap[0] != '+') 348 return; 349 } else if (ap[1] == '-') 350 return; 351 if (!isdigit((unsigned char)ap[1])) 352 continue; 353 /* 354 * Digit signifies an old-style option. Malloc space for dash, 355 * new option and argument. 356 */ 357 len = strlen(ap); 358 if ((start = p = malloc(len + 3)) == NULL) 359 err(1, "malloc"); 360 *p++ = '-'; 361 *p++ = ap[0] == '+' ? 's' : 'f'; 362 (void)strcpy(p, ap + 1); 363 *argv = start; 364 } 365 } 366 367 static void 368 usage(void) 369 { 370 (void)fprintf(stderr, 371 "usage: uniq [-c | -d | -D | -u] [-i] [-f fields] [-s chars] [input [output]]\n"); 372 exit(1); 373 } 374